Introduction: Distributions and inference for categorical data -- Describing contingency tables -- Inference for two-way contingency tables -- Introduction to generalized linear models -- Logistic regression -- Building, checking, and applying logistic regression models -- Alternative modeling of binary response data -- Models for multinomial responses -- Loglinear models for contingency tables -- Building and extending loglinear models -- Models for matched pairs -- Clustered categorical data: marginal and transitional models -- Clustered categorical data: random effects models -- Other mixture models for discrete data -- Non-model-based classification and clustering -- Large- and small-sample theory for multinomial models -- Historical tour of categorical data analysis -- Appendix A: Statistical software for categorical data analysis -- Appendix B: Chi-squared distribution value

A classic in its own right, this book continues to provide an introduction to modern generalized linear models for categorical variables. The text emphasizes methods that are most commonly used in practical application, such as classical inferences for two- and three-way contingency tables, logistic regression, loglinear models, models for multinomial (nominal and ordinal) responses, and methods for repeated measurement and other forms of clustered, correlated response data. Chapter headings remain essentially with the exception of a new one on Bayesian inference for parametric models. Other major changes include an expansion of clustered data, new research on analysis of data sets with robust variables, extensive discussions of ordinal data, more on interpretation, and additional exercises throughout the book. R and SAS are now showcased as the software of choice. An author web site with solutions, commentaries, software programs, and data sets is available"